Skip to main content

MSI Tumor Analysis Implicates Lynch Syndrome in Several New Cancer Types

CHICAGO (GenomeWeb) – A new pan-cancer analysis suggests that there may be a benefit to doing germline testing for Lynch syndrome in advanced cancer patients who have high microsatellite instability (MSI) in their tumors. 

At the American Society of Clinical Oncology annual meeting here this week, researchers from Memorial Sloan-Kettering Cancer Center presented data from a prospective study involving 15,045 cancer patients who had their matched tumor and normal samples analyzed on MSK-IMPACT, an FDA-authorized, next-generation sequencing-based pipeline for detecting MSI as well as mutations in hundreds of tumor-associated genes.

After clustering the tumors based on their MSI scores, the team overlaid information on Lynch syndrome gene mutations in the patients' germlines, demonstrating that the autosomal dominant cancer predisposition syndrome appears to be dramatically over-represented in individuals with tumors in the MSI high (MSI-H) and intermediate/indeterminate (MSI-I) group, but rare in individuals with microsatellite stable (MSS) tumors.

Half of the tumors that occurred in the apparent Lynch syndrome individuals with MSI-H or MSI-I tumors were colorectal or endometrial — the cancer types that are best known in Lynch syndrome.

But the remaining tumors came from a range of other types. And many of those occurred in individuals who would not have met the standard clinical criteria for Lynch syndrome testing, explained the study's senior author Zsofia Kinga Stadler, a medical oncologist at Memorial Sloan-Kettering and clinic director for the MSKCC clinical genetics service.

Consequently, she and her colleagues are advocating for broader Lynch syndrome testing in advanced cancer patients with MSI-H tumors — particularly as MSI profiling has become prevalent in advanced or metastatic cancers since the US Food and Drug Administration approved the anti-PD-1 immune checkpoint inhibitor pembrolizumab for treating advanced solid tumors with MSI-H status.

"This would result in an increased ability to recognize Lynch syndrome, not only in our cancer patients, but also in at-risk family members who may benefit from genetic testing for Lynch and subsequent enhanced cancer surveillance and risk reduction measures," Stadler said at an ASCO press briefing that featured the research this weekend.

Commenting on the study during that briefing, ASCO expert Shannon Westin, a gynecologic oncology and reproductive medicine researcher at the MD Anderson Cancer Center, called it an "absolutely practice-changing study."

"MSI not only has therapeutic implications, but it also has cancer prevention implications," Westin said. "We've only been testing the tip of the iceberg of patients that are affected by Lynch syndrome."

MSKCC's Alicia Latham Schwark presented further details on the analysis during an ASCO clinical science symposium today.

The team used MSK-IMPACT and MSIsensor data to cluster tumors into the MSI-H, MSI-I, and MSS groups, Schwark reported, identifying tumor types that appeared to be particularly MSI-prone. MSI-H tumors made up one-quarter of the small bowel cancers considered, for example, while 16 percent and 14 percent of endometrial and colorectal tumors were MSI-H, respectively. A large proportion of adrenocortical carcinomas had an MSI-I status.

After receiving IRB approval to unmask MSK-IMPACT germline data for the Lynch syndrome genes, the team was also able to determine whether those MSI classifications coincided with the presence of pathogenic or likely pathogenic changes in Lynch syndrome-related mutations in the Lynch syndrome-related DNA repair genes MLH1, MSH2, MSH6, PMS2, and EPCAM.

Indeed, more than 16 percent of the 326 tumors classified as MSI-H contained Lynch syndrome-related mutations, as did 1.9 percent of the 699 tumors designated as MSI-I. In contrast, just 0.3 percent of the 14,020 MSS tumors contained the telltale Lynch syndrome mutations in MLH1, MSH2, MSH6, PMS2, or EPCAM — on par with the 0.3 percent Lynch syndrome prevalence rate previously reported in the general population.

As anticipated, a significant proportion of the tumors with Lynch syndrome mutations — 33 of 66 Lynch mutation-positive MSI-H or MSI-I cases — came from individuals with CRC or endometrial cancer. But the remaining tumors did not.

Lynch syndrome mutations turned up in the germline of individuals with prostate cancer, sarcoma, mesothelioma, and other cancer types not linked to the condition in the past. And nearly half of the non-CRC/endometrial cancer cases with Lynch gene mutations and MSI-H or MSI-I status came from cases that would not meet the current criteria for Lynch syndrome testing.

Stadler noted that similar research is needed for early-stage cancer cases, in case opportunities arise to intervene in burgeoning tumors with MSI, for example. "What if there are malignancies that truly have a higher rate of MSI in the early stage? Then, the implication of this could even be bigger," she told GenomeWeb.

Likewise, it remains to be seen whether there might be a benefit to doing enhanced screening for cancer types not previously linked to Lynch syndrome, Stadler said, depending on the penetrance and lifetime risk of these cancers and the availability of reliable screening strategies for them.

The MSI-H tumors were often marked by DNA mismatch repair deficient (MMR-D) mutational signatures identified using the tumor sequence data, the team reported, though MSI and MMR-D status still appeared to provide complementary information on tumors.

From their data, the researchers also began teasing out MSI-H/I prevalence in tumors from individuals carrying germline mutations to specific Lynch syndrome genes. MSI was common in tumors from individuals carrying MSH2 or MLH1 mutations in their germline, for example. That instability was less common in tumors from individuals with germline changes to MSH6 or PMS2, which have a lower cancer risk and higher mutation frequency in the general population.

At this morning's clinical science symposium, the University Hospital Leaven's Sabine Tejpar discussed the potential feasibility of more routine tumor MSI profiling, highlighting the available data and testing strategies available.

There is ongoing debate over the best strategy for identifying the cancer-related mismatch repair deficiencies that can produce MSI — be it MSI PCR, MSI next-generation sequencing, or immunohistochemical staining — Stadler said, adding that the College of American Pathologists and ASCO recently established an expert panel to explore such questions.